Diesel-like hydrocarbon fuels by catalytic cracking of fat, oils, and grease (FOG) from grease traps

Transcription

1 Diesel-like hydrocarbon fuels by catalytic cracking of fat, oils, and grease (FOG) from grease traps H. da Silva Almeida b,c ; O. A. Corrêa a ; C. C. Ferreira a ; H. J. Ribeiro a,c ; D. A. R. de Castro a,c ; M. S. Pereira c,f ; A. de Andrade Mâncio a,c ; M. C. Santos a,c ; S. A. P da Mota d ; J. A. da Silva Souza a,c ; Luiz E. P. Borges f ; N. M. Mendonça b ; N. T. Machado a,c,e a Laboratory of Separation Processes and Applied Thermodynamic (TERM@) 1 Faculty of Chemical Engineering-UFPA b Laboratory of Multiuser Water and Sludge Analysis and Treatment (LAMAG) b Faculty of Environment and Sanitary Engineering-UFPA c Graduate Program of Natural Resource Engineering-UFPA Rua Augusto Corrêia Nº. 1, CEP: , Cx. P. 8619, Belém-Pará-Brazil d Faculty of Materials Engineering-UNIFESSPA Quadra 17, Bloco 4, Lote Especial, Nova Marabá, CEP: , Marabá-Pará-Brazil e Leibniz-Institüt für Agrartechnik Potsdam-Bornin e.v, Department of Postharvest Technology Max-Eyth-Allee 100, Potsdam 14469, Germany f Laboratory of Catalyst Preparation and Catalytic Cracking 3 Section of Chemical Engineering-IME Praça General Tibúrcio Nº. 80, CEP: , Rio de Janeiro-RJ-Brazil Abstract This work aims to investigate systematically the catalytic cracking of residual fat, oils, and grease (FOG) from grease traps to produce heavy diesel-like hydrocarbon fuels. The experiments were carried out in batch mode at 450 ºC and 1.0 atmosphere, with 5, 10, and 15% (wt.) Na2CO3 using a laboratory scale borosilicate-glass reactor of 143 ml, and with 10% (wt.) Na2CO3 in a pilot scale stirred tank slurry reactor of 143 L. The organic liquid products (OLP) were physical chemistry analyzed for acid and saponification values, density, kinematic viscosity, refractive index, flash point, and copper strip corrosion. FT-IR provided the qualitative chemical composition of OLP and heavy diesel-like fraction. The chemical compositions of OLP and heavy diesel-like hydrocarbon fuel determined by RMN and GC-MS. The laboratory experimental data showed an OLP yield ranging from to 65.40% (wt.), a

2 coke yield ranging between 7.50 and 9.80% (wt.), and a gas yield ranging from to 22.06% (wt.), while pilot scale data showed an OLP yield of 66.57% (wt.), a coke yield of 9.73% (wt.), and a gas yield of 18.24% (wt.). In addition, the yield of OLP obtained in laboratory scale shows a slight increase with Na2CO3 content, while the density, kinematic viscosity, acid value, and saponification value decreases in an exponential fashion with Na2CO3 content. The OLP acid values varied from to mg KOH/g, the density between and g/cm³, and the kinematic viscosity from 3.90 to 6.35 mm 2 s -1 for laboratory data, while in pilot scale, the OLP acid value was mg KOH/g, density of g/cm³, and the kinematic viscosity of mm 2 s -1. The yield of heavy diesel-like hydrocarbon fraction average 19.35% (wt.) with an acid value of 7.27 mg KOH/g, density of g/cm³, and kinematic viscosity of 3.51 mm 2 s - 1. FT-IR, RMN, and GC-MS analysis of OLP and heavy diesel-like hydrocarbon fraction confirms the presence of functional groups characteristic of hydrocarbons (alkenes, alkanes, ring-containing alkenes, and ring-containing alkanes, cycloalkenes, cycloalkanes, and aromatics) and oxygenates (carboxylic acids, ketones, fatty alcohols, and dienes). The GC-MS analysis of OLP and heavy diesel-like hydrocarbon fraction obtained in pilot scale with 10% (wt.) Na2CO3 identified in OLP 76.97% hydrocarbons (39.44% alkenes, 31.91% alkanes, 4.12% ring-containing alkenes, and 1.50% ring-containing alkenes) and 23.03% oxygenates (12.14% carboxylic acids, 6.98% ketones, 1.90% fatty alcohols, and 2.01% dienes). The heavy diesellike hydrocarbon fuel is composed by 69.27% (area) hydrocarbons (32.04% alkenes, 25.62% alkanes, 3.93% ring-containing alkenes, 0.98% ring-containing alkanes, 1.35% cycloalkanes, 4.88% cycloalkanes, and 0.47% aromatics) and 31.75% (area) oxygenates (4.80% carboxylic acids, 4.85% ketones, and 21.08% fatty alcohols). The physical-chemistry properties and chemical composition of heavy diesel-like hydrocarbon fuel obtained by catalytic cracking of residual fat, oils, and grease (FOG) from grease traps with 10% (wt.) Na2CO3 shows the process is technically feasible.

3 Keywords: Catalytic Cracking; Grease Traps; Distillation, Green Diesel, Process Performance. Corresponding author INTRODUCTION Food services generate wastewaters streams containing fats, oils, and grease (FOG) and suspended solids [1-7]. The direct discharge of fats, oils, and grease-containing wastewaters into public sewers may causes reduction of sanitary sewers performance and flowing capacity [5, 7], as well as clogging in drain pipes and/or sewer lines [2, 7], and corrosion of sewer lines due to anaerobic reactions [2]. The management of wastewater streams containing fats, oils, and greases (FOG) poses a set of technical challenges (logistic, pre-treatment, transformation, etc.), particularly in medium and large cities of developing countries, [2, 5, 7]. Despite the proposal of processes and devices for removal of fats, oils, and grease (FOG) from grease-containing wastewater, reported in the literature [1, 5, 7-10], such as grease trap filters [1, 5], ultrafiltration [9], and the use of microorganism [10], the most common pretreatment for fats, oils, and grease (FOG) removal from grease-containing wastewater is a grease trap [5, 7]. In this context, based on the lipid nature of fats, oils, and greases (FOG), processes have been proposed for converting this residual fat material into biofuels [4, 6, 11-14], as well as to improve biogas production from anaerobic digesters at wastewater treatment units [7. 15, 16]. The residual fat material from grease trap, a lipid-base material of low quality, consists of fatty acids, frying oils (soybean, sunflower, etc.), animal fats, hydrogenated fats, fatty alcohols, and other compounds [4, 8, 14, 17]. A process that makes it possible the use of oils and fats of low quality for producing liquid and gaseous fuels is pyrolysis [18-23], and/or catalytic-cracking [17, 24-39].

4 Conversion of low quality lipid materials (fats, oils, and grease) can be achieved either by pyrolysis [18-23] or by catalytic cracking [17, 24-39]. Application of pyrolysis includes used sunflower oil [18], waste fish oil [19, 20], waste frying oil [21, 22], and industrial fatty wastes [23]. The catalytic cracking of low quality lipid material includes fats, oils, and grease (FOG) [17], frying oil [24], used sunflower oil [25], used palm oil and palm oil fatty acid mixture [26-31], used vegetable oil [32], fatty acids and animal fats [33-37], meat and bone meal [38], and the residues of rendering plants [39]. Almeida et al. [17] investigated the production of liquid fuels by catalytic cracking of residual fats, oils, and grease (FOG) in using a stirred tank slurry reactor of 143 L, operating in batch mode. The experiments carried out at 450 C, 1 atm, with 5, 10, and 15% (wt.) activated red mud. The catalytic cracking of residual fat, oils, and grease (FOG) showed an OLP yield ranging from to 75.92% (wt.) with acid values between and mg KOH/g and kinematic viscosity between and mm 2 s -1. The experiment with 5% (wt.) activated Red Mud showed a light diesel yield of 6.39% (wt.) with an acid value of mg KOH/g, and a heavy diesel yield of 41.33% (wt.) with an acid value of mg KOH/g. Dandik and Aksoy investigated the pyrolysis of used sunflower oil, as well as the effect of catalyst on the catalytic cracking of used sunflower oil with Na2CO3 [18, 25]. The experiments were carried out using a laboratory scale reactor (ID = 45 mm, L = 210 mm, VR = 334 cm³), coupled to fractionation columns of different heights (ID = 45 mm, L1 = 180 mm, L2 = 360 mm, and L3 = 540 mm). The columns packed with ceramic rings. Dandik and Aksoy [18] studied the pyrolysis and catalytic cracking of residual sunflower oil with different catalysts (Na2CO3, silica-alumina and HZSM-5), using a laboratory scale reactor coupled to a fractionation packed column of 540 mm length, obtaining gaseous and liquid products (hydrocarbons, carboxylic acids, water), coke, and residual oil. The results showed conversions up to 73.17% (wt.) and OLP yield of 32.80% (wt.) using Na2CO3 as catalyst at 420 C. The use

5 of Na2CO3 also produced OLP with hydrocarbons in the temperature boiling range of gasoline. Dandik and Aksoy [25] studied the catalytic cracking of residual sunflower oil, coupled to fractionation packed column with different heights, at 400 and 420 C, with 1, 5, 10, and 20% (wt.) Na2CO3, obtaining conversions between 43 and 83% (wt.). The OLP composition is strongly dependent on temperature and Na2CO3 content. In addition, increasing the Na2CO3 content and temperature, increased OLP and gas yields and decreased the yield of aqueous phase, acid phase and coke residual oil. By increasing the column length, the amount of gas and coke residual oil increases and that of liquid hydrocarbon and acid phase decreases. The highest OLP yield of 36.4% (wt.) obtained by using 10% Na2CO3 and a packed column of 180 mm at 420 C. Meier et al. [19, 21, 22] investigated the fast pyrolysis of waste fish oil [19], and waste cooking oil [21, 22], using a continuous pilot reactor (ID = 70 mm, L = 2310, VR = 8890 ml). Fast pyrolysis of waste fish oil was carried out at 525 C and mass flow rate of 3 kg/h [19]. The yield of bio-oil ranged from 72 to 73% (wt.), producing after distillation, bio-oil fractions within the temperature boiling ranges of gasoline, light and heavy diesel fuels. The bio-oil fractions were physical chemistry characterized and compared to Diesel S10 specification of ANP N 65 [40], showing physical-chemistry properties and compositional characteristics similar to petroleum fuels. Fast pyrolysis of waste frying oil were carried out at 475, 525 and 575 C, 50% (wt.) water, and mass flow rate of 3 kg/h [21]. The yields of OLP varied between 56 and 77% (wt.), yields of coke ranged from 0 to 17% (wt.), and that of gas from 20 to 44% (wt.). The GC analysis identified the presence of hydrocarbon fraction between C4-C10 and C9-C16. Fast pyrolysis of waste frying oil carried out at 475, 500, and 525 C, and mass flow rates between 0.78 to 3.65 kg/h [22]. The yields of OLP varied between 49.5 and 75.2% (wt.), with acid values between and [mg KOH/g]. The yield of light bio-oil was 78.4% (wt.) at 525 C

6 and mass flow rate of 0.78 kg/h. The yield of heavy bio-oil was 52.4% (wt.) at 475 C and mass flow rate of 2.86 kg/h. Takwa Kraiem et al. [20] investigated the pyrolysis of waste fish fats at 500 C, 5 C min -1, and N2 flow rate of 0.3 cm³ min -1, using a fixed bed bench scale reactor (ID = 150 mm, L = 300 mm, VR = 5300 ml). The yield of OLP was 54.60% (wt.), formed by 37.50% (wt.) bio-oil and 17.10% (wt.) of aqueous phase. The bio-oil presented an acid value of mg KOH/g, ph of 3.12, and kinematic viscosity (40 C) of mm 2 s -1. Santos et al. [23] investigated the pyrolysis of industrial fatty waste (soybean soap stock, beef tallow and poultry industry waste) to produce hydrocarbons. The pyrolysis of industrial fatty wastes (soybean soap stock, beef tallow, and poultry industry waste), produced OLP consisting of hydrocarbons and oxygenated compounds. Chemical analysis of diesel-like fractions showed the presence of olefins, paraffins, carboxylic acids and esters. The yields of OLP ranged from 8.0 to 32.0% (wt.), being the maximum obtained with beef tallow. The results showed acid values of 3.02 (soybean soap stock), (beef tallow), and (chicken fat) mg KOH/g, densities from to g/cm³, and kinematic viscosities from 3.02 to 4.93 mm 2 s -1. Buzetzki et al. [24] investigated the catalytic cracking of filtered frying oils in the presence of 10% (wt.) zeolite (NaY, Clinoptilolite-CL, and HZSM-5), using a laboratory scale tubular neck reactor of 400 cm³ with mechanical stirring in batch mode. The experiments carried out at the temperature interval of C, heating rate of 10 C/min, and 1.0 atm, showed that yield of untreated condensates (OLP + Aqueous Layer) ranged from 85 to 93% (wt.), the yield of coke ranged from 3 to 7% (wt.), while that of non-condensable gaseous products ranged from 4 to 9% (wt.). The acid values of treated condensates, after separation of aqueous layer and evaporation of short carboxylic acids, ranged between 74 and 116 mg KOH/g, the densities between and g/cm³, while the kinematic viscosities

7 between and mm 2 s -1. The conversion of acyl esters into hydrocarbons was incomplete, as confirmed by the presence of carboxylic acids in OLP in high concentrations. Ooi et al. [26-30] studied the catalytic cracking of palm oil fatty acid mixtures, fatty acid mixtures, palm oil fatty acids, and used palm oil. The experiments were carried out using a fixed-bed micro scale reactor of cm³, at temperature between C and 1.0 atm, weight hourly space velocity of h -1, over HZSM-5, MCM-41/ zeolite and a mixture of zeolite and MCM-41, mesoporous catalysts (AlMCM-41 and LPMM-41), composite catalysts (HZSM-5 and MCM-41/ZSM-5), microporous HZSM-5, and mesoporous MCM- 41/SBA-15 molecular sieve. For the catalytic conversion of fatty acids mixture into liquid fuels, LPMM-41 produced a gasoline yield of 43% (wt.) after distillation of OLP [28], while application of microporous HZSM-5 and mesoporous MCM-41/SBA-15 molecular sieve at 450 C and weight hourly space velocity of 2.5 h -1, showed conversions up 98% (wt.) and a gasoline yield 44% (wt.) [30]. W. H. Chang and C. T. Tye [31], investigated the catalytic cracking of used palm oil using composite zeolites (H-ZSM5, Cu-HZSM5, Zn-HZSM5, Mg-HZSM5), at 350 C and 1.0 atm, using a batch laboratory scale reactor of 1000 cm³, with mechanical stirring. The obtained OLP, gas and residue yields ranged between and 84.70% (wt.), 5.63 and 15.80% (wt.), and 8.20 and 10.81% (wt.), respectively. The best results was obtained with Mg-HZSM5, producing OLP, gas and residue yields of 84.70% (wt.), 5.63 % (wt.), and 9.64 % (wt.), respectively. In addition, Mg-HZSM5 produced the highest gasoline and diesel fractions of 9% (vol.) and 72.5% (vol.), respectively. Charusiri and Vitidsant [32], studied the conversion of used frying oil into liquid fuels, at the temperature between 400 and 430 C, reaction time between 30 and 90 min, and H2 pressure between 10 and 30 bar, over sulfated zirconia, using a 70 cm³ batch micro scale reactor. The optimum conditions were obtained at 430 C, 90 min, 10 bar, producing the highest conversion of gasoline ( 24.38%), as well as kerosene, light gas oil,

8 gas oil, residues, hydrocarbon gases, and small amounts of solids ( 11.98%, 24.35%, 5.70%, 13.86%, 19.07%, and 0.65%), respectively. Weber et al. [33, 35], investigated the catalytic conversion of fatty acid mixture (60:40 oleic and stearic acids) and animal fat [33], as well as the chemical composition of hydrocarbons obtained by catalytic cracking of animal fats [35], using a continuous moving bed reactor in pilot scale (L = 6.0 m, B =2.4 m, H = 3.1 m). The catalytic cracking of fatty acids mixture (60:40 oleic to stearic acids) carried out at 410 and 450 C, mass flow rate of 10 kg/h, using Na2CO3 as catalyst and 5% (wt.) H2O, while the mass flow rate of animal fat varied between 10 and 40 kg/h [33]. For fatty acids conversion, the bio-oil yield varied from 64 to 74% (wt.) with an acid value between 0.64 and 0.80 mg KOH/g, while the bio-oil yield of animal fat conversion ranged from 60 to 70% (wt.) with an acid value between 0.5 and 1.80 mg KOH/g, and mean kinematic viscosity of 1.78 mm 2 s -1. The yield of gaseous products ranged from 25 to 30% (wt.), and that of coke between 4 and 6% (wt.). The distillation of bio-oil yields 66% (wt.) bio-diesel and 21% (wt.) bio-gasoline [33]. The thermo-chemical conversion of animal fat carried out at 430 C, with approximately 10.0% (wt.) Na2CO3 and 5% (wt.) H2O, in order to produced salts of carboxylic acids [35]. The chemical analysis of bio-oil identified alkanes and alkenes. In addition, distillation of bio-oil yields diesel-like fractions matching nearly all parameters of European Diesel Fuel standards [35]. Liew et al [36], investigated the catalytic cracking of waste chicken with zeolite catalyst (ZSM-5), at 400 C, using a laboratory scale distillation apparatus under N2 flow, obtaining a bio-oil containing hydrocarbons chain length from C7 C24, as well as aliphatics, carboxylic acids, alcohols, ketones, esters, aromatics, anhydrides, ether, and aldehydes. Hua Tian et al. [37] studied the catalytic cracking of palm and soybean oils, as well as chicken fat, over zeolites (USY and ZSM-5), using a laboratory scale two-stage riser fluid catalytic cracking unit. Experiments were carried out at 1.0 atm, 500 C (1 st stage riser), and 520 C (2 nd stage riser),

9 catalyst to feed mass ratios of 6 and 8 and residence time of 1.4 and 1.7 s, respectively. The results showed a liquid yield (LPG + gasoline + diesel) of 78.49% (wt.), an LPG yield of 34.34% (wt.), a gasoline yield of 32.75% (wt.), a diesel yield of 11.40% (wt.), a dry gas yield of 4.48% (wt.), a heavy oil yield of 2.95% (wt.), and a coke yield of 2.31% (wt.). The acid value of diesel fraction was 190 mg KOH/ml and the kinematic 3.91 mm² s -1. Weber et al. [38] and Bojanowski et al. [39] investigated the thermal and catalytic conversion of animal meal and MBM (animal meal, meat and bone meal) at 400 C, using a bench scale Pyrex glass reactor of 140 cm length and 7 cm internal diameter (VReactor 5400 ml). The experimental results for the pyrolysis of animal meal showed an OLP (bio-oil) yield of 29.2% (wt.), a solid (coke + salt) yield of 46.8% (wt.), a water yield of 14% (wt.), and a gas yield of 10% (wt.), with a density of 0.94 g/ml and kinematic viscosity of 45.2 mm²/s. The pyrolysis of MBM showed an OLP yield of 16.3% (wt.), density of 0.90 g/ml and kinematic viscosity of 29.8 mm²/s. RMN analysis of OLP obtained from MBM identified the presence of 97% (area.) aliphatic and 3% (area.) aromatic compounds, although FT-IR analysis of MBM also confirms the presence of carboxylic acids and acyl derivatives (esters of triglycerides). The catalytic cracking of animal fat and animal meal, meat and bone meal was carried out in a vertical bench-scale quartz glass reactor (ID = 29 mm, L = 1100 mm), with a 30 cm catalyst bed, connected above a cracking vessel of 1000 ml [34, 38, 39]. Granular (2 5 mm) ZSM-5 zeolite-type catalysts such as Y-type zeolite (e.g. Wessalith) and H-type zeolite (e.g. Pentasil) were investigated at 400 [34, 39], and H-type zeolite (e.g. Pentasil) at 550 C [34], and 8.7% (wt.) Na2CO3 with 5% (wt.) H2O at C [39]. The experimental results for catalytic conversion of animal fat at 400 C showed an OLP (bio-oil) yield of 72.90% (wt.), coke yield of 5.80% (wt.), and gas yield of 16.20% (wt.), with density of 0.80 g/ml and kinematic viscosity of 2.90 mm²/s for Y-type zeolite. OLP (bio-oil) yield of 56.74% (wt.), coke yield of 0.87% (wt.), and gas yield of % (wt.) for H-type zeolite at 400 C. RMN analysis of OLP

10 obtained from animal fat indicates a chemical composition similar to that of MBM. The experimental data for thermal catalytic conversion of animal fat at 550 C showed an OLP (biooil) yield of 31.5% (wt.), with a density of 0.84 g/ml and kinematic viscosity of 0.74 mm²/s. The experimental results for the catalytic cracking of animal meal with 8.7% (wt.) Na2CO3, 5% (wt.) H2O between C, showed an OLP (bio-oil) yield of 21.61% (wt.), a solid (coke + salt) yield of 10.42% (wt.), a water yield of 11.24% (wt.), with an acid value of 13.3 mg KOH/g and kinematic viscosity of 14.5 mm²/s [39]. Rectification of bio-crude from animal meal produces 65.8% (wt.) bio-diesel and 13.3% (wt.) bio-gasoline. RMN and GC-MS analysis confirm the presence of alkylbenzenes as major compounds. FT-IR analysis indicates the presence of carboxylic acids and acyl esters. This work aims to investigate systematically the technical feasibility of producing heavy diesel-like fuels by catalytic cracking of residual fats, oils, and grease (FOG) from grease traps at 450 ºC and 1.0 atmosphere, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale, and 10% (wt.) Na2CO3 in a pilot scale in batch mode MATERIALS AND METHODS 2.1. Materials The residual fat, oils, and grease (FOG) used as renewable raw material, was collected at the grease traps of Students Restaurant (UFPA). Solvay Chemicals International SA (Brussels, Belgium) supplied Sodium Carbonate (commercial soda ASH Light D50) with 98.0% (wt.) purity Pre-treatment of residual fat, oils, and grease material (FOG) The pre-treatment of residual fat material, a complex mixture consisting of residual fat, oils, and grease (FOG), aqueous phase and suspended solids is described in details elsewhere[17].

11 Physical-chemistry analysis of residual fat, oils, and grease material (FOG) The dehydrated residual fat material (FOG) has been physical-chemistry characterized for acid value (AOCS Cd 3d-63), saponification value (AOCS Cd 3-25), free fatty acids (ASTM D5555), density (ASTM D1480), and kinematic viscosity (ASTM D445) using a Cannon- Fenske viscometer (Schot Geräte, Model: ) with a capillary tube Nº. 200 [17] Catalyst characterization Na2CO3 used as catalyst has been analyzed by XRF to determine its purity grade. In a previous study the mineralogical and qualitative chemical analysis of Na2CO3 has been investigate by XRD, TGA/DTG, and FT-IR [41] Characterization of OLP and heavy diesel-like fraction Physical-chemistry analysis of OLP and heavy diesel-like fraction OLP and heavy diesel-like fraction, obtained by fractional distillation of OLP produced with 10% (wt.) Na2CO3 in pilot scale, have been physical chemistry characterized according to the analysis described in section 2.3. In addition, OLP and heavy diesel-like fraction were analyzed for corrosiveness to copper (ASTM D130-12) using a copper strip test analyzer (PETROTEST DP, Model: E ), flash point (ASTM D93) using a Pensky-Martens close cup analyzer (TANAKA, Model: APM-7/FC-7), and refractive index (AOCS Cc 7-25) [17] FT-IR analysis of OLP and heavy diesel-like fraction OLP produced with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale, as well as heavy diesel-like fraction, obtained by fractional distillation of OLP produced with 10% (wt.) Na2CO3 in pilot scale, were analyzed by FT-IR to provide the qualitative chemical composition of OLP and heavy diesel-like fraction, being the analytical procedures described in details elsewhere [17]. FT-IR analysis of residual fat, oils, and grease (FOG) was described in a previous study [17].

12 RMN analysis of OLP and heavy diesel-like fraction The NMR 1 H and 13 C spectra of OLP and heavy diesel-like fraction were obtained using a RMN spectrometer (Varian, Model UNITY 300), with a resonance frequency of 300 MHz. The solvent was deuterated chloroform (CDCl3) and as a reference standard substance [(CH3)4Si] was used, being the acquisition conditions of 1 H NMR spectra obtained with a pulse at 30 and 32 transients, while that of 13 C NMR spectra obtained with a pulse at 30 and 3940 transients GC-MS Analysis of PLO and Kerosene-like Fractions Prior to the chemical analysis by GC-MS, derivatization of fatty acids present in OLP and diesel-like fraction was carried out as described elsewhere [17]. The GC-MS analysis of OLP and heavy diesel-like fraction was performed as described in a previous study [17]. 2.6 Experimental Apparatus and Procedures Laboratory Scale Unit Figure 1 describes the bench scale experimental apparatus. It consists of a cylindrical borosilicate-glass reactor of 150 mm height and 35 mm diameter with a 34/35 standard taper outer joint connected to a 300 mm jacket length borosilicate glass condenser unit with 24/40 standard taper outer and inner joints. A cylindrical heating system constructed of ceramic resistances of 800 W with a digital temperature control (Therma, Model: TH90DP ) and a thermocouple of type K (Ecil, Model: QK-2), with a precision of 1.0 K jacket the borosilicate-glass reactor. The upper part of the borosilicate-glass adapter with a 34/35 standard taper inner joint connected to the reactor and a 24/40 standard taper outer joint to the condenser unit, consisting of a small size glass tube was connected to a N2 flow stream (carrier gas for the gaseous reaction products obtained during the catalytic cracking process), by using a capillary silicone tube from nitrogen gas reservoir. A heating plate with a magnetic stirrer (Ika, C-MAG,

13 Model: HS7), provided agitation with aid a magnetic bar inside the borosilicate-glass reactor. A glass sampling unit of 250 ml, a cylindrical nitrogen reservoir with a two stage pressure relieve valve (Cemper, Model: CS_54), a gas flow meter (Omel, Model: ), calibrated with air at 1 atmosphere and 294 K, within the range ml/min, and a gas exhaust system release the non-condensable gases into the atmosphere. Initially, approximately 50 ml of sample is weighed in a semi-analytical balance (QUIMIS, Model: Q 500L210C), with g accuracy. Afterwards, 10% (wt.) of Na2CO3 is weighed. Then, the solid paste dehydrated residual fat was liquefied using a heating plate with a magnetic stirrer (IKA, C-MAG, Model: HS7) at 150 C to remove all the remaining water. Afterwards, the dehydrated residual fat material, the catalyst, and the magnetic bar were introduced inside the glass borosilicate reactor. Then, the reactor was connected to the condenser cooled with water at 10 C, being inserted inside the cylindrical furnace (heating system) and connected to a nitrogen cylinder. The operating temperature was set to 450 C with a heating rate of 10 C/min, and a nitrogen flow to 40 ml/min. The reaction time was the time necessary to reach the reactor set point temperature and the cracking temperature defined as the temperature the gaseous products were formed. The collected liquid product was placed within a separation funnel in order to remove the aqueous phase, being the organic liquid products washed (liquid-liquid extraction) three times with distilled water in the proportion 2:1 at 70 C. The samples were stored in vials for subsequent physical-chemistry analysis. Figure 1: Laboratory scale borosilicate-glass catalytic-cracking reactor of 143 ml Pilot Scale Unit The apparatus described in details elsewhere [41], operates in batch mode at 450 ºC and 1.0 atmosphere, using residual fat, oils, and grease material (scum) from grease traps as renewable raw material and 10% (wt.) Na2CO3 as catalyst. Almeida et al. [17], described in

14 details the experimental apparatus and procedures of fractional distillation of OLP to obtain diesel-like fractions. Figure 2: Pilot scale carbon steel catalytic-cracking reactor of 143 L. 2.7 Material Balance of Thermal Catalytic Cracking Process The steady state mass balance within the stirred tank sludge bed reactor, operating at atmospheric pressure and batch mode, as well as yield of reaction products was described in a previous study [17]. The dehydrated residual fat, oils, and grease (FOG) material used as renewable raw material on the catalytic cracking experiments in laboratory and pilot scales presented acid values between and mg KOH/g. Thus, conversion of fatty acids was defined in terms of acid values of residual fat, oils, and grease (FOG) and OLP, as computed by equation (1). WhereIA is the acid value of dehydrated residual fat material, and IAOLP the acid value of OLP. RF IA IA IA RF RF OLP Conversion of FFA % (1) 3 RESULTS AND DISCUSSIONS 3.1 Physical-chemistry analysis of dehydrated fats, oils, and grease (FOG) Table 1 presents the physical-chemistry analysis of residual fat, oils, and grease (FOG) after pre-treatment described elsewhere [17], used as renewable raw material on the thermal catalytic cracking experiments in laboratory and pilot scales. It can be observed that dehydrated residual fat material used on the catalytic cracking experiments in laboratory scale is free of moisture and has an acid value of mg KOH/g, showing that residual fat, oils, and grease (FOG) material is composed mainly by carboxylic acids. This is probably due to hydrolysis reactions as well as microorganism degradation of acyl esters, confirmed by the free fatty acids

15 content of 77.71% (wt.). The residual fat material used on the catalytic cracking experiment in pilot scale still contains residual water with an acid value of mg KOH/g, showing that residual fat, oils, and grease (FOG) material is a lipid rich material under microorganism decomposition containing fatty acids, waste-frying oils, as well as animal and butter fats, confirmed by the free fatty acids content of 36.26% (wt.) and saponification value of mg KOH/g. Table 1: Physical-chemistry characterization of dehydrated residual fats, oils, and grease (FOG) used as renewable feed on the catalytic creaking experiments at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale Characterization of NA2CO XRF analysis XRF of Na2CO3 has been evaluated to determine the main oxides, the chemical elements present on its crystalline phases, but mainly its purity grade. Table 2 shows the quantitative chemical composition of Na2CO3 by XRF. It could be observed the presence of oxides (Al2O3, SiO2, CaO, and Fe2O3), and Na in high concentration with purity of 97.83% (wt.). Table 2: XRF of Na2CO SEM analysis Figure 3 illustrates the scanning electron microscopy of Na2CO3. Soda ASH Light D50 (Na2CO3) particles exhibit geometry similar with a cylindrical and/or a rectangular shape. The images are in accord to the micrographics obtained by Forryan et al. [42], who reported that the cylindrical shape of particles provides a more realistic representation of Na2CO3 morphology. Figure 3: SEM of Na2CO Catalytic cracking of residual fats, oils, and grease (FOG) Process conditions and material balances of catalytic cracking of fats, oils, and

16 grease (FOG) in laboratory and pilot scales The process conditions, material balances, and yields of reaction products (OLP, Coke, Gas, and H2O) obtained by catalytic cracking of residual fat, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale, and 10% (wt.) Na2CO3 in pilot scale are shown in Table 3 and Figure 4. For the catalytic cracking experiments in laboratory scale, the initial cracking temperature decreases in a linear fashion with increasing Na2CO3 content. In addition, there has been observed a great difference between initial cracking temperatures in laboratory (T = 362 C) and pilot (T = 306 C) scales. This is due to the high performance thermal insulation system of catalytic cracking pilot unit, constructed with special refractory bricks, which maintains the external reactor wall temperature at 50 C, thus avoiding heat losses to the environment. In addition, the pilot unit mechanical impeller system has a higher power and torque compared to the small magnetic bar (L = 10 mm) used to provide agitation inside the laboratory scale borosilicate-glass reactor of 143 ml. The experimental data illustrated in Table 3 presents OLP yields ranging from to 66.57% (wt.), showing a percentage relative error of 1.75% between OLP yields in pilot and laboratory scales with 10% (wt.) Na2CO3. The yield of coke ranged from 7.50 to 9.73% (wt.), showing a percentage relative error of 0.79% between coke yields in pilot and laboratory scales with 10% (wt.) Na2CO3. The yield of gas ranged from to 22.06% (wt.), showing a percentage relative error of 10.63% between gas yields in pilot and laboratory scales with 10% (wt.) Na2CO3, while the yield of H2O varied between 5.46 and 8.50% (wt.), showing a percentage relative error of 35.76% between H2O yields in pilot and laboratory scales with 10% (wt.) Na2CO3. The results obtained for OLP yields in laboratory and pilot scale are according to similar studies reported in the literature [17, 18, 24, 25, 31-39]. particularly to those of Weber et al. [33, 35], who investigated the production of liquid fuels by catalytic cracking of a mixture

17 from fatty acids and animal fat using Na2CO3 as catalyst and Almeida et al. [17], who studied the catalytic cracking of residual fats, oils, and grease from grease traps with activated Red Mud as catalyst. The yield of coke ranged from 7.50 to 9.73% (wt.), being in accord to similar studies reported in the literature [17, 24, 31-33, 35-36, 38-39]. The yield of gaseous products ranged from to 22.06% (wt.). The results are according to similar studies reported in the literature [17, 24, 31-33, 35-39]. The yield of H2O ranged from 5.46 to 8.50% (wt.). The results are according to similar studies reported in the literature [17, 24, 34, 38-39, 41]. Figure 4 illustrates the yields of reaction products (OLP, Gas, Coke, and H2O) obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. One may observes that OLP yields varies between and 66.57% (wt.), with a slight sigmoidal tendency to increase with increasing Na2CO3 content, while the yields of coke and H2O show a slight sigmoidal tendency to decrease with increasing Na2CO3 content. Table 3: Process parameters and overall steady state material balances of catalytic creaking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. Figure 4: Yields of reaction products (OLP, Coke, Gas, and Water) obtained by thermalcatalytic cracking of dehydrated residual fat, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale Physical-chemistry characterization of OLP Figures 5, 6, and 7 and Table 4 illustrate the physical-chemistry properties of OLP obtained by catalytic cracking of fats, oils, and grease (FOG) from grease traps at 450 C and

18 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory and 10% (wt.) Na2CO3 in pilot scale. The experimental results in laboratory scale show that acid values, saponification values, densities, and kinematic viscosities of OLP ranged from to mg KOH/g, to mg KOH/g, to g/cm³, and to mm 2 s -1, respectively. The acid value, saponification value, density, and kinematic viscosity of OLP in pilot scale are equal to mg KOH/g, mg KOH/g, g/cm³, and mm 2 s -1, respectively. The OLP acid values are higher compared to Weber et al. [33], Bojanowski et al. [39], and Mota et al. [41], but lower than similar studies reported in the literature [17, 22-23, 37]. The OLP densities values are lower compared to similar studies reported in the literature [17, 24, 38, 39], and in accord to Santos et al. [23], Bojanowski et al. [39], and Mota et al. [41]. The OLP kinematic viscosities are lower compared to similar studies reported in the literature [17, 20, 24, 38, 39], higher than the values reported by Weber et al. [33] and Mota et al. [41], and according to Santos et al. [26] and Hua Tian et al. [37]. Figures 5, 6, and 7 show that density, kinematic viscosity, as well as acid and saponification values of OLP decreases in a smooth exponential fashion with increasing Na2CO3 content. Na2CO3 has not only proven its high performance and selectivity to convert large amounts of fatty acids and acyl esters into hydrocarbons, but also to decrease the acid value of OLP, a fundamental physical-chemistry parameter conferring the quality of biofuels. In addition, the conversion of fatty acids computed by equation (1) ranged from to 86.86%, and increases with increasing Na2CO3 content. The measured OLP physical (density and kinematic viscosity) and physical chemistry (acid and saponification values) properties are lower compared to those reported by Almeida et al. [17], showing that Na2CO3 is highly selective and effective to thermochemical convert low quality fats, oils, and grease (FOG) into hydrocarbons than activated Red Mud. Table 4: Physical-chemistry characterization of OLP obtained by catalytic cracking of

19 dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. Figure 5: Density of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. Figure 6: Kinematic viscosity of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. Figure 7: Acid and Saponification values of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale Fractional distillation of OLP Material balances and yields of fractional distillation of OLP Table 5 shows the mass balances and yields (Distillates and Raffinate) of fractional distillation of OLP obtained by thermal-catalytic cracking of dehydrated residual fat, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. It has not been observed the production of gasoline-like fractions, showing that Na2CO3 was not selective for hydrocarbon fractions in the temperature boiling point range of gasoline (40 C < T < 175 C), which is according to the results of Weber et al. [35]. However, it was observed in all distillation fractions of PLO the presence of hydrocarbon fractions in the temperature boiling point range of kerosene (175 C < T Boiling < 235 C), light diesel (235 C < T Boiling < 305 C) and heavy diesel (305 C < T Boiling < 400 C) fuels. The yields of kerosene, light diesel, and heavy diesel-like fuels were 14.90% (wt.), 32.01% (wt.), and 19.35% (wt.), respectively, while the yield of raffinate was 32.25% (wt.), being the global yield of biofuels

20 equal to 66.26% (wt.). The results are lower compared to Bojanowski et al. [39], higher compared to Dandik and Aksoy [25] and Hua Tian et al. [37], and in accord to data obtained in pilot and laboratory scale reported elsewhere [17, 32, 41]. Table 5: Mass Balances and Yields (Distillates and Raffinate) of fractional distillation of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale Physical-chemistry properties of heavy diesel-like fraction Heavy diesel-like fraction (305 < T Boiling < 400 C) of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale were physical-chemistry characterized and the results shown in Table 6. Physical-chemistry analysis presented in Table 6 shows that except for the density, which is lower than the lower density limit of diesel S10 specification of ANP 65 [40], kinematic viscosity and flash point matches the diesel S10 specification of ANP 65 [40]. This is probably due to the presence of light diesel compounds C13-C17 in heavy diesellike fraction, caused by limitations of fractional distillation apparatus (Vigreux Column of 03 Stages), as discussed in Section 3.6. The results are in accord to similar data obtained in pilot scale reported elsewhere [17, 40]. Table 6: Physical-chemistry characterization of heavy diesel-like fraction (305 C < T Boiling < 400 C) obtained by fractional distillation in laboratory scale of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. 3.4 FT-IR Analysis of OLP and heavy diesel-like fraction FT-IR analysis of OLP Figures 8 and 9 illustrate the FT-IR analysis of OLP obtained by catalytic cracking of

21 dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory, and 10% (wt.) Na2CO3 in pilot scale, respectively. The identification of absorption bands/peaks was done according to previous studies [17, 23, 41]. The spectra of OLPs produced with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale exhibit intense peaks at 2923 cm -1 and 2852 cm -1, indicating the presence of aliphatic compounds, associated to methylene (CH2) and methyl (CH3) groups. This confirms the presence of hydrocarbons [17, 23, 41]. It can also be seen the presence of an intense axial deformation band for the OLPs obtained with 15% (wt.) Na2CO3 in laboratory and 10% (wt.) Na2CO3 in pilot scales, characteristic of carbonyl (C=O) groups, between cm -1, being the peaks at 1717 cm -1 probably associated to ketones and/or carboxylic acids [17, 23, 41]. In addition, it could be observed that the axial deformation bands between cm -1 are not only broader but also the peaks intensity at 1713 cm -1 are higher for the organic liquid products produced with 5 and 10% (wt.) Na2CO3 in laboratory scale. This is probably due to higher OLP acid values, as described in Table 4. The spectra of organic liquid products obtained with 5 and 10% (wt.) Na2CO3 in laboratory scale presents an intense and larger axial deformation band between cm -1, compared to the spectra of organic liquid products obtained with 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. These bands indicates the presence of a hydroxyl (O-H) group, characteristics of carboxylic acids [17, 23, 41], presented in higher concentrations in the organic liquid products obtained with 5 and 10% (wt.) Na2CO3 in laboratory scale, as described in Table 4. The spectra of OLPs produced with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale exhibits at 1460 and 1461 cm -1, a characteristic asymmetrical deformation vibration of methylene (CH2) and methyl (CH3) groups, indicating the presence of alkanes [17, 23, 41]. The spectrum of OLP produced with 10% (wt.) Na2CO3 in pilot scale identify at 1380 cm -1, bands of symmetrical angular deformation of C-H bonds in methyl group (CH3) [17, 41]. The peaks

22 at 992, 966, 906 and 909 cm -1, observed in the organic liquid products obtained with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale are characteristic of an angular deformation outside the plane of C-H bonds, indicating the presence of alkenes [17, 23, 41]. The spectra of OLPs obtained with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale exhibits bands between 719 and 722 cm -1, characteristic of an angular deformation outside the plane of C-H bonds in methylene (CH2) group, indicating the presence of olefins [17, 23, 41]. The analysis of OLPs spectra produced with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale identify the presence of aliphatic groups (alkenes, alkanes, etc.), as well as oxygenates (carboxylic acids, ketones, etc.). Figure 8: FT-IR of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale. Figure 9: FT-IR of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale FT-IR analysis of heavy diesel-like fraction Figure 10 illustrates the FT-IR spectra of heavy diesel-like fraction (305 C < T Boiling < 400 C) obtained by fractional distillation of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. The identification of absorption bands/peaks was done according to previous studies [17, 23, 41]. The spectrum of heavy diesel-like fraction exhibits bands between 2958 and 2922 cm -1, and 2852 and 2958 cm -1, with intense peaks at 2922 cm -1, 2958 cm -1 and 2852 cm -1, associated to methylene (CH2) and methyl (CH3) groups, indicating the presence of aliphatic compounds (hydrocarbons) [17, 23, 41]. The presence of a peak at 1717 cm -1, characteristic of carbonyl (C=O) groups, is probably associated to ketones and/or carboxylic acids [17, 23, 52]. In addition, it was observed a broad axial deformation band between 3200-

23 cm -1, indicating the presence of a hydroxyl (O-H) group, characteristics of carboxylic acids [17, 23, 41]. The peak at 1461 cm -1, characteristic of an asymmetrical deformation vibration of methylene (CH2) and methyl (CH3) groups, indicates the presence of alkanes [17, 23, 41]. The spectrum also presents a peak at 1380 cm -1, characteristic of symmetrical angular deformation of C-H bonds in methyl group (CH3) [17, 41]. In addition, the spectrum exhibits peaks at 992, 966 and 909 cm -1, characteristics of an angular deformation outside the plane of C-H bonds, indicate the presence of alkenes [17, 23, 41]. The peak at 722 cm -1 is characteristic of an angular deformation outside the plane of C-H bonds in methylene (CH2) group, indicating the presence of olefins [17, 23, 52]. The FT-IR analysis of heavy diesel-like fraction (305 C < T Boiling < 400 C) identify the presence of aliphatic groups (alkenes, alkanes, etc.), as major chemical compounds, as well as oxygenates (carboxylic acids, ketones, etc.). Figure 10: FT-IR of OLP distillation fractions (kerosene-like, light diesel-like, and heavy diesel-like fractions) obtained by thermal-catalytic cracking of dehydrated residual fat, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. 3.5 RMN analysis of OLP and heavy diesel-like fraction The 13 C and 1 H RMN spectra of OLP and heavy diesel-like fraction produced by fractional distillation of OLP obtained by catalytic cracking of dehydrated residual fat, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale are illustrated in Figures 11, 12, 13, and 14. The 13 C RMN spectra of OLP and heavy diesellike fraction identified chemical shifts characteristic of R-CH2 (methylene) and R-CH3 (methyl) groups of linear carbon chain between ppm and ppm, for OLP and heavy diesel-like fraction, respectively, indicating the presence of aliphatic hydrocarbons. The 13 C RMN spectra also exhibits signals characteristic of olefins with chemical shifts of carbon double bonds (C=C) observed at and ppm for both OLP and kerosene-like

24 fraction. The 1 H RMN spectrum shows chemical shifts of hydrogen bounds to carbon with peaks between ppm and ppm, for OLP and heavy diesel-like fraction, respectively, indicating the presence of aliphatic hydrocarbons, as well as chemical shifts of hydrogen and methine (=CH-) group with peaks between ppm and ppm, for OLP and heavy diesel-like fraction, respectively. The chemical shifts between ppm and ppm, for OLP and heavy diesel-like fraction, respectively, indicate the presence of hydrogen bonded to an unsaturated carbon of olefins in OLP and heavy diesel-like fraction. Figure 11: 13 C RMN spectra of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. Figure 12: 1 H RMN spectra of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. Figure 13: 13 C RMN of heavy diesel-like fraction of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. Figure 14: 1 H RMN of heavy diesel-like fraction of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. 3.6 GC-MS analysis of OLP and heavy diesel-like fraction Figures 15 and 16 illustrates the chromatograms of OLP and heavy diesel-like fraction obtained by catalytic cracking of dehydrated residual fat, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. Table 7 describes the class

25 of compounds, the summation of peak areas for a class of compounds, and retention times of chemical compounds identified by CG-MS. The GC-MS analysis identified approximately 60 compounds, classified into two major groups (hydrocarbons and oxygenates). The main classes of hydrocarbons detected by GC-MS were alkenes, alkanes, ring-containing alkenes, and ringcontaining alkanes, cycloalkenes, cycloalkanes, and aromatics while the main classes of oxygenates were carboxylic acids, ketones, fatty alcohols, and dienes. The OLP presents on its composition 75.97% (area) hydrocarbons and 24.03% (area) oxygenates. The hydrocarbon were composed by 39.44% alkenes (area), 30.91% alkanes (area), 1.50% ring-containing alkanes (area), and 4.12% ring-containing alkenes (area), while the oxygenates were composed by 12.14% carboxylic acids (area), 6.98% ketones (area), 1.90% fatty alcohols (area), and 2.01% dienes (area). The concentration (area) of carboxylic acids identified by GC-MS in OLP is according to the OLP acid value of [mg KOH/g]. This confirms that Na2CO3 was not only effective but also selective to thermo-chemical convert fatty acids and lipid base compounds present in fat, oils, and grease (FOG) into liquid hydrocarbons, according to studies reported in the literature [18, 22, 33, 41]. GC-MS analysis of OLP identified the presence of carboxylic acids, including C15H32O (DCA), a product of chlorination of drinking water, produced by the microbiological metabolism of chlorine-containing compounds such as NaClO, a common disinfectant in food services, with 6.84% (area). Hexadecanoic and (9Z)- Octadecenoic acids, the dominating carboxylic acids in soybean and palm oils, as well as hydrogenated fats, were also present in OLP with 3.08 and 2.22% (area), respectively. The heavy diesel-like fractions of PLO obtained by catalytic cracking of dehydrated residual fat, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale presents on its composition % (area) hydrocarbon and 31.75% (area) oxygenates. The hydrocarbon were composed by 32.04% (area) alkenes, 25.62% (area) alkanes, 3.93% (area) ring-containing alkenes, 0.98% (area) ring-containing alkanes, 1.35% (area)

26 cycloalkanes, 4.88% (area) cycloalkenes, and 0.47% (area) aromatics, while the oxygenates were composed by 4.80 % carboxylic acids (area), 4.85% (area) ketones, and 21.08% (area) fatty alcohols. The concentration of carboxylic acids identified by GC-MS in heavy diesel-like fraction is consistent to its acid value of 7.27 [mg KOH/g]. The hydrocarbons identified in OLP by GC-MS present carbon chain length ranging from C10 to C22 with following carbon chain lengths, alkenes C10-C19, alkanes C10-C22, ring-containing alkenes C10-C13, and ring-containing alkanes C11-C15, indicating the presence of heavy gasoline compounds with C10 (C5-C10), kerosene-like fractions (C11-C12), light diesel-like fractions (C13-C17), and heavy diesel-like fractions (C18-C25), as reported in the literature [57]. The heavy diesel-like fraction presents carbon chain lengths ranging from C12 to C23 with following carbon chain lengths, alkenes C14-C23, alkanes C13-C22, ringcontaining alkenes C12-C14, ring-containing alkanes C15, cycloalkenes C13, cycloalkanes C12-C20, and aromatics with C12. It may be observe the presence of light diesel compounds C13-C17 in heavy diesel-like fraction. This is probably due to the limitation of fractional distillation apparatus (Vigreux Column of 03 Stages). The presence of light diesel compounds C13-C17 in heavy diesel-like fraction has probably contribute to a deviation of 0.97%, compared to the lower density limit specification for diesel fuel S10 of ANP 65 [40]. Despite a deviation of 0.97% compared to the lower density limit specification for diesel fuel S10 of ANP 65 [40], the heavy diesel-like fraction matches the ANP 65 for viscosity, flash point, and copper strip corrosion. Figure 15: GC-MS of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale.

27 Figure 16: GC-MS of heavy diesel-like fraction of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. Table 7: Class of compounds, summation of peak areas, and retention times of chemical compounds identified by CG-MS of OLP and heavy diesel-like fraction obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. 4 CONCLUSIONS For the catalytic cracking experiments in laboratory scale, the initial cracking temperature decreases in a linear fashion with increasing Na2CO3 content. The OLP yields varies between and 66.57% (wt.) for the catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. The results show a percentage relative error of 1.75% between OLP yields in pilot and laboratory scales with 10% (wt.) Na2CO3. In addition, the OLP yield shows a slight (sigmoidal) tendency to increase with increasing Na2CO3 content, while the yields of coke and H2O show a slight (sigmoidal) tendency to decrease with increasing Na2CO3 content. Experimental data for the catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale show that density, kinematic viscosity, as well as acid and saponification values of OLP decreases in a smooth exponential fashion with increasing Na2CO3 content. This confirms the high performance and selectivity of Na2CO3 to convert large amounts of fatty acids and acyl esters into hydrocarbons, making it possible to decrease the acid value of OLP, a fundamental physical-chemistry parameter conferring the quality of biofuels. The FT-IR analysis of OLPs obtained with 5, 10, and 15% (wt.) Na2CO3 in laboratory

28 scale and 10% (wt.) Na2CO3 in pilot scale, and heavy diesel-like fraction (305 C < T Boiling < 400 C) identify the presence of aliphatic groups (alkenes, alkanes, etc.), as major chemical compounds, as well as oxygenates (carboxylic acids, ketones, etc.), in agreement with RMN and GC-MS analysis. The GC-MS analysis of OLP obtained in pilot scale with 10% (wt.) Na2CO3 presents on its composition 76.97% hydrocarbons and 23.03% oxygenates. The heavy diesel-like fraction is composed by % (area) hydrocarbon and 31.75% (area) oxygenates. The hydrocarbons were composed by 32.04% (area) alkenes, 25.62% (area) alkanes, 3.93% (area) ring-containing alkenes, 0.98% (area) ring-containing alkanes, 1.35% (area) cycloalkanes, 4.88% (area) cycloalkenes, and 0.47% (area) aromatics, while the oxygenates were composed by 4.80 % carboxylic acids (area), 4.85% (area) ketones, and 21.08% (area) fatty alcohols. Physicalchemistry analysis of heavy diesel-like fuel shows that except for the density, which is lower than the lower density limit of diesel S10 specification of ANP 65 [40], kinematic viscosity and flash point matches the diesel S10 specification of ANP 65 [40]. The results show that catalytic cracking of fats, oils, and grease (FOG) scum from grease at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale is technically feasible ACKNOWLADGMENT ELETROBRÁS S/A for the Project financial support. The first author would like to express his gratitude to Prof. Dr. Luiz E. P. Borges (Examiner), for introducing the author into the marvelous field of catalysis and thermal catalytic cracking of vegetable oils, and to Prof. Dr.-Ing Nélio T. Machado (Supervisor), for giving the opportunity to Research as Assistant at the Laboratory of Thermal Separation Processes and Applied Thermodynamics. 691

29 REFERENCES [1] Aziz, T., Holt, L., Keener, K., Groninger, J., Ducoste, J. Performance of Grease Abatement Devices for Removal of Fat, Oil, and Grease. J. Environ. Eng., Vol. 137, Issue 1 (2011), [2] Uwe Stoll, Hani Gupta. Management Strategies for Oil and Grease Residues. Waste Management & Research January (1997) 15, 23-32l [3] Eddy, N. Restaurants and commercial facilities present specific problems for onsite systems. Small Flow J., 12 (2), 1998, 1-5 [4] Canakci, M. The potential of restaurant waste lipids as biodiesel feedstocks. Bioresource Technology 98 (2007), [5] N. H. Wong, P. L. Law, S. H. Lai. Field tests on a grease trap effluent filter. Int. J. Environ. Sci. Tech., 4 (3), 2007, [6] Zhong-Ming Wang, Jin-Suk Lee, Ji-Yeon Park, Chuang-Zhi Wu, Zhen-Hong Yuan. Optimization of biodiesel production from trap grease via acid catalysis. Korean J. Chem. Eng., 25(4), 2008, [7] J. Hunter Long, Tarek N. Aziz, Francis L. de los Reyes III, Joel J. Ducoste. Anaerobic codigestion of fat, oil, and grease (FOG): A review of gas production and process limitations. Process Safety and Environmental Protection 90 (2012), [8] Iman A. F. Husain, Ma an Fahmi Alkhatib, Mohamed Saedi Jammi, Mohamed E. S. Mirghani, Zaki Bin Zainudin, Asif Hoda. Problems, Control, and Treatment of Fat, Oil, and Grease (FOG): A Review. Journal of Oleo Science. Vol. 63 (2014), N 8, [9] Robert A. Dangel, David Astraukis, Jean Palmateer. Fatty Acid Separation from Hydrolyser Wastewater by Ultrafiltration. Environmental Progress, 14(1), (1995) [10] N. G. Wakelin, C. F. Forster. An investigation into microbial removal of fats, oils and greases. Bioresource Technology, 59(1), (1997)

30 [11] Canakci, M., Van Gerpen, J Biodiesel from oils and fats with high free fatty acids. Transaction of the American Society of Agricultural Engineering, 44, [12] Walterio Diaz-Felix, Mark R. Riley, Werner Zimmt, Michael Kazz. Pretreatment of yellow grease for efficient production of fatty acid methyl esters. Biomass and Bioenergy, Vol. 33, Issue 4, April 2009, [13] Montefrio, M. J., Xinwen, T., Obbard, J.P Recovery and pre-treatment of fats, oil and grease from grease interceptors for biodiesel production. Applied Energy 87, [14] Parichart Hasuntree, Vichuta Toomthong, Sukanda Yoschoch, Usarat Thawornchaisit. The potential of restaurant trap grease as biodiesel feedstock. Songklanakarin J. Sci. Technol. 33 (5), , 2011 [15] Andres Donoso-Bravo, Maria Fdz-Polanco. Anaerobic co-digestion of sewage sludge and grease trap: Assessment of enzyme addition. Process Biochemistry, Vol. 48, Issues 5 6, 2013, [16] Kabouris, J.C., Tezel, U., Pavlostathis, S.G., Englemann, M., Dulaney, J., Gillette, R.A., Todd, A.C., 2009b. Methane recovery from the anaerobic co-digestion of municipal sludge and FOG. Bioresource Technology, 100 (15), (2009) [17] H. da Silva Almeida, O. A. Corrêa, J. G. Eid, H. J. Ribeiro, D. A. R. de Castro, M. S. Pereira, L. M. Pereira, A. de Andrade Mâncio, M. C. Santos, J. A. da Silva Souza, Luiz E. P. Borges, N. M. Mendonça, N. T. Machado. Production of Biofuels by Thermal Catalytic Cracking of Scum from Grease Traps in Pilot Scale. Journal of Analytical and Applied Pyrolysis, Available online 8 January [18] Dandik, L., Aksoy, H. A. Effect of catalyst of used Oil Carried Out in a Fractionating Pyrolysis reactor. Renewable Energy 16 (1999)

31 [19] V. R. Wiggers, A. Wisniewski Jr., L. A. S. Madureira, A. A. Chivanga Barros, H. F. Meier. Biofuels from waste fish oil pyrolysis: Continuous production in a pilot plant. Fuel, Vol. 88, Issue 11, 2009, [20] Takwa Kraiem, Aïda Ben Hassen-Trabelsi, Slim Naoui, Habib Belayouni, Mejdi Jeguirim. Characterization of the liquid products obtained from Tunisian waste fish fats using the pyrolysis process. Fuel Processing Technology, Vol. 138, October 2015, [21] H. F. Meier, V. R. Wiggers, G. R. Zonta, D. R. Scharf, E. L. Simionatto, L. Ender. A kinetic model for thermal cracking of waste cooking oil based on chemical lumps. Fuel, Vol. 144, 2015, [22] V. R. Wiggers, G. R. Zonta, A. P. França, D. R. Scharf, E. L. Simionatto, L. Ender, H. F. Meier. Challenges associated with choosing operational conditions for triglyceride thermal cracking aiming to improve biofuel quality. Fuel, Volume 107, May 2013, [23] Andre L.F. Santos, Danilo U. Martins, Osvaldo K. Iha, Rafael A.M. Ribeiro, Rafael L. Quirino, Paulo A.Z. Suarez. Agro-industrial residues as low-price feedstock for diesel-like fuel production by thermal cracking. Bioresource Technology, Vol. 101, Issue 15, 2010, [24] E. Buzetzki, K. Sidorová, Z. Cvengrošová, J. Cvengroš. Effects of oil type on products obtained by cracking of oils and fats. Fuel Processing Technology. 92 (2011) [25] Dandik, L., Aksoy, H. A. Pyrolysis of used sunflower oil in the presence of sodium carbonate by using fractionating pyrolysis reactor. Fuel Processing Technology. Vol.57 (2), 81-92, 1998 [26] Ooi Y. S, Ridzuan. Zakaria, Abdul Rahman Mohamed and Subhash Bhatia. Catalytic conversion of palm oil-based fatty acid mixture to liquid fuel. Biomass and Bioenergy, Vol. 27, Issue 5, 2004,

32 [27] Ooi Y. S., Zakaria R., Mohamed A. R., Bhatia S. 2004b. Synthesis of Composite Material MCM-41/Beta and Its Catalytic Performance in Waste Used palm Oil Cracking. Applied Catalysis A: General, Volume 274, Issues 1 2, 2004, [28] Ooi Y. S., Zakaria R., Mohamed A. R., Bhatia S. Catalytic Conversion of Fatty Acids Mixture to Liquid Fuels over Mesoporous Materials. Reaction Kinetics and Catalysis Letters, Vol. 84, Issue 2, , 2005 [29] Ooi Y. S., Zakaria R., Mohamed A. R., Bhatia S. 2004c. Catalytic Cracking of Used Palm Oil and Palm Oil Fatty Acids Mixture for the Production of Liquid Fuel: Kinetic Modeling. J. Am. Chem. Soc. 18: [30] Ooi Y. S., Zakaria R., Mohamed A. R., Bhatia S. 2005b. Catalytic Conversion of Fatty Acids Mixture to Liquid Fuel and Chemicals over Composite Microporous/Mesoporous Catalysts. Energy Fuels, 2005, 19 (3), [31] W. H. Chang and C. T. Tye. Catalytic Cracking of Used Palm Oil Using Composite Zeolite. The Malaysian Journal of Analytical Sciences, Vol 17 No 1 (2013), [32] Witchakorn Charusiri and Tharapong Vitidsant. Kinetic Study of Used Vegetable Oil to Liquid Fuels over Sulfated Zirconia. Energy Fuels, 2005, 19 (5), [33] B. Weber, E. A. Stadlbauer, S. Stengl, M. Hossain, A. Frank, D. Steffens, Elmar Schlich, Gerhard Schilling. Production of hydrocarbons from fatty acids and animal fat in the presence of water and sodium carbonate: reactor performance and fuel properties. Fuel, 94 (2012), [34] E. A. Stadlbauer, R. Altensen, S. Bojanowski, G. Donnevert, A. Fiedler, S. Hossain, J. Rossmanith, G. Shilling. Herstellung von Kohlenwasserstoffen aus Tierfett durch thermokatalytisches Spalten, Erdöl Erdgas Kohle 2006, 122 (2), 64 69

33 [35] B. Weber, E. A. Stadlbauer, S. Eichenauer, A. Frank, D. Steffens., S. Elmar, S. Gerhard. Characterization of Alkanes and Olefins from Thermo-Chemical Conversion of Animal Fat. Journal of Biobased Materials and Bioenergy, Vol. 8, N 5, 2014, (12) [36] M. Y. Liew, A. Salmiaton, W. A. K. G. Wan Azlina, R. Yunus. Catalytic Pyrolysis of Waste Chicken Fats Using Zeolite Catalysts. Developments in Sustainable Chemical and Bioprocess Technology, Springer US ISNB: [37] Hua Tian, Chunyi Li, Chaohe Yang, Honghong Shan. Alternative Processing Technology for Converting Vegetable Oils and Animal Fats to Clean Fuels and Light Olefins. Chinese Journal of Chemical Engineering, Vol. 16, Issue 3, 2008, [38] S. Eichenauer, B. Weber, E. A. Stadlbauer. Thermochemical Processing of Animal Fat and Meat and Bone Meal to Hydrocarbons based Fuels. ASME 2015, 9 th International Conference on Energy Sustainability, Paper N ES , V001T02A001, doi: /ES [39] S. Bojanowski, A. Fiedler, A. Frank, E. A. Stadlbauer, G. Schilling. Catalytic Production of Liquid Fuels from Organic Residues of Rendering Plants. Eng. Life Sci. 2007, 7, Nº 6, [40] ANP Nº 65, DOU [41] S. A. P. Mota; A. A. Mancio, D. E. L. Lhamas, D. H. de Abreu, M. S. da Silva, W. G. dos Santos, D. A. R. de Castro, R. M. de Oliveira, M. E. Araújo, L. E. P. Borges, N. T. Machado. Production of green diesel by thermal catalytic cracking of crude palm oil (Elaeis guineensis Jacq) in a pilot plant. Journal of Analytical and Applied Pyrolysis. 110 (2014) 1 11 [42] C. L. Forryan, R. G. Compton, O. V. Klymenko, C. M. Brennan, C. L. Taylorc, M. Lennon. Comparative solubilisation of potassium carbonate, sodium bicarbonate and sodium carbonate in hot dimethylformamide: application of cylindrical particle surface-controlled dissolution theory. Phys. Chem. Chem. Phys. 8 (2006) [43] W. L. Nelson. Petroleum Refinery Engineering. 4 th Edition, McGraw-Hill, 1985

34 Table 1: Physical-chemistry characterization of dehydrated residual fats, oils, and grease (FOG) used as renewable feed on the catalytic creaking experiments at 450 C and 1.0 atm, with 5, 10. and 15% (wt.) Na 2CO 3 in laboratory scale and 10% (wt.) Na 2CO 3 in pilot scale. Physical-Chemistry Analysis Feed Feed Hasuntree et. al [14] Kanacki Kanacki (Laboratory Scale) (Pilot Scale) [4] SM-01 [4] SM-08 Na 2CO 3 [wt.%] Na 2CO 3 [wt.%] ρ 60 C [g/dm³] C [cst] C [cp] Moisture [%] Insoluble Solids [%] Acid Value [mg KOH/g] Saponification Value [mgkoh/g] Peroxide Value [meq of O 2/kg] Unsaponifiable Matter [%] Ester Index [mg KOH/g] FFA [%] SIM-01: Restaurant grease, partially processed. The solids and free water were removed but it had not been cooked. SIM-08: Unprocessed restaurant grease. Collected from the tops of three separate barrels. Water mostly at the bottom of the barrel. Ester Index = Saponification Value Acid Value

35 825 Table 2: XRF of Na2CO Oxides/Chemical Elements Na2CO3 [wt.%] Al2O SiO CaO Fe2O P Na Cl Ag Ʃ

36 Table 3: Process parameters and overall steady state material balances of catalytic creaking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na 2CO 3 in laboratory scale and 10% (wt.) Na 2CO 3 in pilot scale. Process Parameters Na 2CO 3 [wt.%] (Laboratory Scale) Na 2CO 3 [wt.%] (Pilot Scale) 10 Cracking Temperature [ C] Mass of Hydrated Residual Fat [g] Mass of Na 2CO 3 [g] Cracking Time [min] Mechanical Stirrer Speed [rpm] Initial Cracking Temperature [ C] Mass of H 2O (V) (T = 100 C, 1.0 atm) [g] Mass of Liquid Phase (OLP+H 2O) [g] Mass of Coke [g] Mass of OLP [g] Mass of H 2O [g] Mass of Gas [g] Yield of Coke [wt.%] Yield of OLP [wt.%] Yield of H 2O [wt.%] Yield of Gas [wt.%]

37 Table 4: Physical-chemistry characterization of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale. Physical-Chemistry Properties OLP Na2CO3 [wt.%] (Pilot Scale) OLP Na2CO3 [wt.%] (Laboratory Scale) ANP Nº 65 ρ [g/cm 3 ] Acid Value [mg KOH/g] IARF IAOLP [%] - IA RF 858 Refractive Index [-] μ [cst] Flash Point [ C] Saponification Value [mg KOH/g] Ester Index [mg KOH/g] Copper Strip Corrosion (1A) ANP: Brazilian National Petroleum Agency, Resolution N 65 (Specification of Diesel S10)

38 Table 5: Mass Balances and Yields (Distillates and Raffinate) of laboratory scale fractional distillation of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. Distillation: Vigreux OLP [g] Gas [g] Raffinate [g] Distillates [g] Yield [wt.%] Boiling Temperature Column of 03 Stages K LD HD K LD HD T I [ C] T F [ C] 10 [wt.%] Na 2CO T I : Initial Boiling Temperature, T F : Final Boiling Temperature, K = Kerosene, LD = Light Diesel, HD = Heavy Diesel

39 Table 6: Physical-chemistry characterization of heavy diesel-like fraction (305 C < T Boiling < 400 C) obtained by fractional distillation in laboratory scale of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale. Physical-chemistry Properties Na2CO3 10 [wt.%] ANP Nº 65 Heavy Diesel [g/cm 3 ] I. A [mg KOH/g] 7.27 Note I. R[-] Note [cst] I. S [mg KOH/g] Note Ester Index [mg KOH/g] FP [ºC] C (1 A) I.A=Acid Value, I.R=Refractive Index, I.S=Saponification Value, FP=Flash Point, C= Copper Corrosiveness

40 Table 7: Class of compounds, summation of peak areas, and retention times of chemical compounds identified by CG-MS of OLP and heavy diesel-like fractions obtained by fractional distillation in laboratory scale of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na 2CO 3 in pilot scale. OLP, 10% [wt.] Na 2CO 3 Heavy Diesel-like Fraction, 10% [wt.] Na 2CO 3 Class of Compounds: RT [min] Class of Compounds: RT [min] Chemical Compounds Chemical Compounds Alkenes Alkenes 1-Decene Tetradecene Undecene Pentadecene Tetradecene Octadecene Dodecene 6.82 Heptadecene Tridecene Eicosene Pentadecene Eicosene Heptadecene Heptadecene Icosene Nonadecene Heptadecene Tricosene Nonadecene Ʃ (Area.%) = Ʃ (Area.%) = Alkanes Alkanes n-tridecane 7.85 n-decane 4.54 n-hexadecane 8.99 n-undecane Dodecane 9.75 n-dodecane 6.77 Heptadecane n-tridecane 7.85 n-nonadecane Hexadecane 8.99 Heneicosane n-hexadecane Docosane Octadecane Ʃ (Area.%) = Heptadecane Ring containing-alkenes Docosane Heptyl-1-cyclopentene 9.69 Ʃ (Area.%) = Octyl-1-cyclohexene 11.19

41 Ketones Ʃ (Area.%) = Heptadecanone Ring containing-alkanes 2-Pentadecanone ,3-Dicyclohexylpropane Nonadecanone Ʃ (Area.%) = 0.98 Ʃ (Area.%) = 6.98 Cycloalkenes Carboxylic Acids Cyclohexene 9.88 Dichloroacetic acid Ʃ (Area.%) = 1.35 Hexadecanoic acid Cycloalkanes (9Z)-Octadecenoic acid Nonylcyclopropane 6.68 Ʃ (Area.%) = Cyclotetradecane 9.64 Ring containing-alkenes Cyclopentadecane Butylcyclohexene 5.27 Cyclohexadecane Heptyl-1-cyclohexene Cycloeicosane Ʃ (Area.%) = 1.50 Cyclopropane Ring containing-alkanes Ʃ (Area.%) = ,2-Dibutylcyclopropane 5.73 Aromatics 1-Pentyl-2-propylcyclopropane 5.83 Naphthalene Nonylcyclopentane 9.65 Ʃ (Area.%) = ,3-Dicyclohexylpropane Ketones Ʃ (Area.%) = Heptadecanone Fatty Alcohols Ʃ (Area.%) = 4.85 Cis-9-octadecen-1-ol Carboxylic Acids 1-Heptadecanol Dichloroacetic acid Ʃ (Area.%) = 1.90 Tetradecanoic acid Dienes Hexadecanoic acid ,19-Icosadiene Ʃ (Area.%) = 4.80 Ʃ (Area.%) = 2.01 Fatty Alcohols n-tetradecanol Heptadecin-1-ol Hexadecene-1-ol 12.25

42 Octadecenol Hexadecanol Pentadecanol Heptadecanol Ʃ (Area.%) =

43 Figure 1: Laboratory scale borosilicate-glass catalytic-cracking reactor of 143 ml

44 Figure 2: Pilot scale carbon steel catalytic-cracking reactor of 143 L

45 Figure 3: SEM of Na2CO

46 70 Yied of Reaction Products [wt.%] Coke, Y = /[1+10^((12.5-Na Coke 2 CO 3 )*-0.5)] OLP, Y = /[1+10^((7.5-Na OLP 2 CO 3 )*0.5)] H O, Y = /[1+10^((12.5-Na 2 H2O 2 CO 3 )*-0.5)] Gas, Y = *Na Gas 2 CO 3, r² = OLP, Y = (Pilot Scale) OLP Na 2 CO 3 [wt.%] Figure 4: Yields of reaction products (OLP, Gas, Coke, and H2O) obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale

47 0,87 0,86 10% [wt.] Na 2 CO 3 (Laboratory Scale) 10% [wt.] Na 2 CO 3 (Pilot Scale) 0,85 OLP [g/cm³] 0,84 0,83 0, , Na 2 CO 3 [wt.%] Figure 5: Density of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale

48 6,5 6,0 10% [wt.] Na 2 CO 3 (Laboratory Scale) 10% [wt.] Na 2 CO 3 (Pilot Scale) 5,5 OLP [wt.%] 5,0 4,5 4,0 3,5 3, Na 2 CO 3 [wt.%] Figure 6: Kinematic viscosity of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale

49 Acid Value [mgkoh/g] Acid Value (Laboratory Scale) Saponification Value (Laboratory Scale) Acid Value (Pilot Scale) Saponification Value (Laboratory Scale) Saponification Value [mgkoh/g] Na 2 CO 3 [wt.%] Figure 7: Acid and Saponification values of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale and 10% (wt.) Na2CO3 in pilot scale

50 60 OLP 5% [wt.]na 2 CO T[%] OLP 10% [wt.] Na 2 CO 3 60 OLP 15% [wt.]na 2 CO [cm -1 ] Figure 8: FT-IR of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 5, 10, and 15% (wt.) Na2CO3 in laboratory scale. 1020

51 60 10% [wt.] Na 2 CO 3 50 T [%] [cm-1] Figure 9: FT-IR of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale

52 Heavy Diesel 10% Na 2 CO 3 60 T [%] [cm -1 ] Figure 10: FT-IR of heavy diesel-like fraction obtained by fractional distillation in laboratory scale of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale.

53 Intensity Chemical Shift (ppm) Figure 11: 13 C RMN spectra of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale

54 Intensity Chemical Shift (ppm) Figure 12: 1 H RMN spectra of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale

55 Intensity Chemical Shift (ppm) Figure 13: 13 C RMN of heavy diesel-like fraction obtained by fractional distillation in laboratory scale of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale

56 Intensity Chemical Shift (ppm) Figure 14: 1 H RMN of heavy diesel-like fraction obtained by fractional distillation in laboratory scale of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale

57 Figure 15: GC-MS of OLP obtained by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale

58 Figure 16: GC-MS of heavy diesel-like fraction obtained by fractional distillation in laboratory scale of OLP produced by catalytic cracking of dehydrated residual fats, oils, and grease (FOG) from grease traps at 450 C and 1.0 atm, with 10% (wt.) Na2CO3 in pilot scale